EVALUATION OF FIELD PERFORMANCE OF PREFABRICATED … EVALUATION-OF-FIELD... · Bergado et al....
Transcript of EVALUATION OF FIELD PERFORMANCE OF PREFABRICATED … EVALUATION-OF-FIELD... · Bergado et al....
1
EVALUATION OF FIELD PERFORMANCE OF
PREFABRICATED VERTICAL DRAINS (PVD) FOR
SOIL GROUND IMPROVEMENT IN THE
SOUTHERN VIETNAM
Hoang-Hung Tran-Nguyen 1
, Huong Ho 2
, and Hy Hoan Ha 3
Department of Civil Engineering, Ho Chi Minh City University of Technology, Ho Chi Minh City, Vietnam, Tel: +84838637003, +84838650714, e-mail: [email protected].
2, 3Department of Civil Engineering, Ho Chi Minh City University of Technology, Ho Chi Minh City, Vietnam
Received Date: January 8, 2003
Abstract
A 40,000-km2 soft ground area in the southern Vietnam is big obstruction for infrastructure
development in Vietnam. PVD is the most popular technique applied to improve soft ground in the
southern Vietnam. However, field performance of PVD based on field monitoring is quite
discrepant from prediction. This study aims at better understanding of PVD field performance via
case studies. The two complete projects using PVD for soft ground improvement were utilized for
back analysis. Field monitored settlements agree well with predicted settlements when the
field settlement has experienced a settlement of 0.3 m or less. The predicted settlement exceeds the
field monitored settlement about 50% when the field settlement reaches 0.5 m or larger. The
result indicates that degree of filed consolidation was less than 90%, and large field settlement
keeps occurring during the service stage of highway embankments and may cause field
settlement over the allowable settlement according to the Vietnam Code for soft ground
improvement.
Keywords: Consolidation, Highway embankment, Prefabricated vertical drain (PVD), Preload settlement, Soft ground, Wick drain
Introduction
The Southern Vietnam, a 40,000-km2 area, lies on thick soft deposit. This soft deposit causes great difficulty for infrastructure development. Various soft ground improvement technologies have been applied in Vietnam such as vertical drain (sandy drain or wick drain), stone columns, and geotextiles. Prefabricated vertical drain (PVD) is applied widely in Vietnam since 1990s due to its advantages such as low cost, fast installation, safe for environment, and stable supply. However, several projects that PVD was used as a major soft ground improvement technique to improve soft deposit in the South have excess settlement in short time after the projects in service.
Field performance of PVD affects by a number of factors such as deformation, lateral pressure, infiltration, smear zone effects, and so on (Tran-Nguyen et al. 2010, Tran-Nguyen 2010, Tran-Nguyen & Edil 2011). Tran-Nguyen et al. (2010) and Ali (1991) reported that the discharge capacity of PVD reduce significantly when the PVD has experienced a percent settlement of 30% or more. Smear zone which is disturbed zone surrounding a PVD after installation using a mandrel delays horizontal consolidation process due to lower hydraulic conductivity of the smear zone than that of undisturbed zone (Tran-Nguyen & Edil 2011, Tran-Nguyen 2010). PVD discharge capacity decreases remarkably under a lateral pressure of 150 kPa or more (Rixner et al. 1986), that is, with long PVD (e.g., longer 20 m, Hansbo 1981), the PVD performance is affected by lateral pressure.
ASEAN Engineering Journal Part C, Vol 4 No 2 (2015), ISSN 2286-8151 p.9
Bergado et al. (1996b) proposed that PVD discharge capacity should take a factor of
safety of 2 or greater to take infiltration into account.
The Vietnam code for the use of PVD, 22 TCN 244-98 issued by Ministry of Transport
(1998) assumed ideal performance of PVD, that is, all factors affecting the field
performance of PVD are ignored. Consequently, prediction of consolidation settlement
may be different from monitoring field settlement. This paper investigated two complete
projects that used PVD to improve soft ground for back analysis.
Methodology
Prediction of consolidation settlement was analyzed using 22 TCN 244-98 which is the
Hansbo (1979)’s theory and the Finite Element Method (FEM) (Plaxis 2D v.85. software)
to compare with field monitored settlement.
Primary Consolidation Settlement
Consolidation settlement, Sc, is analyzed using Equation (1) and (2) depending
on OCR – over-consolidation ratio.
If OCR = 1,
Sc
Ho
1 e0
oC
clog
vo
vo
(1)
If OCR > 1,
Sc
Ho
1 e0
Cr
log
c
vo
Cc
log
vo
c
(2)
where Ho – thickness of soil in active zone, eo – initial void ratio, Cc – compression index,
Cr - Swelling index, c – pre-consolidation pressure, vo – overburden stress, –
embankment load or surcharge or preload.
Time Settlement by the Hansbo’s Theory (1979)
Consolidation settlement at time t, St, is determined using Equation (3)
St S
c.U
t (3)
where Ut – average horizontal degree of consolidation at time t. By assuming ideal
performance of PVD, Ut can be computed by Equation (4) (Hansbo 1979)
)(/8exp1 nFTU hh (4)
where Th – time factor, Th = Ch.t/D2e, De – diameter of an influence/drainage zone
surrounding a PVD, De =1.13.S, or = 1.05.S, for square or triangular pattern installation,
respectively, S – distance between PVD, Ch – horizontal coefficient of consolidation,
F(n)=ln(n)- 3/4 (Hansbo 1979) - spacing factor, n = De/dw; dw=(a+b)/2 - equivalent
diameter of a PVD (Rixner et al. 1986), a – PVD thickness, b – PVD width.
FEM
In order to utilize the FEM to simulate PVD using the Plaxis 2D version 8.5 software, PVD
needs to be converted into equivalent elements. There several ways to simulate PVD
elements for the 2D FEM (Chai et al. 2001, Indraratna & Redana 2000), and this study
uses two methods for comparison. The Mohr-Coulomb model was applied to simulate the
behavior of the subsoil underneath the highway embankment. The Mohr-Coulomb model
ASEAN Engineering Journal Part C, Vol 4 No 2 (2015), ISSN 2286-8151 p.10
is popular to simulate for soft ground at acceptable accuracy to compare with more robust
soft ground models such as modified Cam Clay model. To increase presision, 2D 15-node
triangular elements were applied for FEM meshes. Considering a PVD as an equivalent
element converted from an axisymmetric model to a plane strain model (Indraratna
& Redana 2000) (FEM-1), and equivalent hydraulic conductivity of a PVD element, khp,
can be defined as Equation (5).
khp
0.67
ln n 0.75
kh
(5)
where kh – horizontal hydraulic conductivity of subsoil.
Alternatively, Chai et al. (2001) proposed a simple method to simulate the whole
subsoil underneath an embankment improved using PVD (FEM-2). They suggested an
equivalent vertical hydraulic conductivity for both the PVD and the subsoil, kve, defined by
Equation (6).
kve 1
2.26.l2
F.De
2
kh
kv
k
v
(6)
where kh, kv – horizontal, vertical hydraulic conductivity of soil surrounding PVD,
respectively, l – vertical drainage length, F = F(n) – spacing factor.
Selected Case Histories
Saigon East-West Highway
Saigon East-West Highway which is a highway with two 50-m-wide embankments for the
two directions crosses Ho Chi Minh City from the East to the West of the City with the
total length of 22 km on soft ground areas (Figure 1). Embankment high of the highway is from 2 to 6 m and several techniques used to improve soft ground for the 22-km-
long highway embankment. A research location is at km 17 + 320 on a section that soft
ground was improved using PVD. The cross-section at this location is shown on Figure
2. Soil properties of the selected location are printed in Table 1. Along the 50-m subsoil
profile at km 17 + 320, there are 3 layers: (1) a 15-m very soft clay layer at the top with
SPT around 0-2 blows, (2) a 6-m medium stiff clay layer, and (3) fine loose sand layer at
the bottom. It is noted that the authors obtained all soil properties from the contractors.
ASEAN Engineering Journal Part C, Vol 4 No 2 (2015), ISSN 2286-8151 p.11
Figure 1. The Saigon east-west highway and study location at km 17 + 320 in Ho Chi
Minh City (Google map)
Chikami PVD (A6) made by Japan (100 x 4 mm) was used at this
location with PVD length of 21.5 m installed in square pattern at a distance of 1.2
m. The PVD discharge capacity was 1000 m3/year.
Figure 2. A selected section at km 17 + 320 on the Saigon East-West highway in Ho Chi
Minh City
Study location at km 17 + 320
End point
Starting point pointlocation at
km 17 + 320
Saigon East - West Highway
ASEAN Engineering Journal Part C, Vol 4 No 2 (2015), ISSN 2286-8151 p.12
Table 1. Soil Properties at the Study Location km 37 + 320 on the Saigon East-West
Highway, Ho Chi Minh City
Layer Thick-
ness γ E v φ c kv kh
m kN/m³ kN/m2
(o) kN/m² m/day m/day
Very Soft clay 15.5 14.5 1160 0.35 9.08 8.8 1.21E-04 3.63E-04
Medium stiff
clay 6.3 15.4 1340 0.3 8.10 8.8 1.47E-05 4.41E-05
Fine loose sand > 15 19.2 1580 0.3 10.30 7.2 2.76E-05 3.10E-05
Field monitored settlement was obtained via the settlement plate E5 located at the
centerline of the embankment (Figure 2). The surcharge was divided into two main stages with several minor surcharge stages between the first and the last surcharge stage.
The first fill height was 1.3 m and maintained in 446 days. Several fill steps were
conducted to raise up to a maximum height of 5.03 m in about 300 days, and the highest fill
was kept until the 1019th
day with the total accumulated settlement of 2.2 m. Figure 3 shows the field monitored accumulated settlement at the study location, km 17 + 320.
-2400
-2000
-1600
-1200
-800
-400
0
400
0 100 200 300 400 500 600 700 800 900 1000
Time (day)
Sett
lem
ent (m
m)
Surc
harg
e(0
.01m
)
Surcharge
Monitored settlement
`
`
Figure 3. Field monitored settlement for 1019 days at the maximum height of 5.03 m.
The settlement varying with time was computed using Hansbo (1979) (or 22 TCN 244-98) (Eqn. 1,2,3,4), FEM-1 (Eqn. 5), and FEM-2 (Eqn. 6). The Mohr-Coulomb modelwas utilized to simulate the subsoil behavior in the Plaxis 2D v8.5 software. Figure 3a andFigure 3b demonstrate FEM meshes using Plaxis 2D v8.5 for FEM-1 and FEM-2respectively, The result is shown in Figure 4.
It can be seen that the computed settlements were significantly diverged from the
monitored settlement at large consolidation settlement (e.g., St >= 0.3 m). In order words,
the assumption of the ideal performance of PVD is not relevant for large consolidation
settlement.
ASEAN Engineering Journal Part C, Vol 4 No 2 (2015), ISSN 2286-8151 p.13
Figure 3a. A FEM-1 mesh for the initial conditions using the PLAXIS 2D software
Figure 3b. A FEM-2 mesh for the initial conditions using the PLAXIS 2D software
Clo
sed c
onso
lidat
ion b
oundar
y
Clo
sed c
onso
lidat
ion b
oundar
y
Closed flow boundary
Closed flow boundary
ASEAN Engineering Journal Part C, Vol 4 No 2 (2015), ISSN 2286-8151 p.14
-2900
-2500
-2100
-1700
-1300
-900
-500
-100
300
0 100 200 300 400 500 600 700 800 900 1000
Time (day)S
ettl
emen
t (m
m)
Su
rch
arg
e (0
.01
m)
.
Surcharge
Monitored settlement
Hansbo 1979 (22TCN 244-98)
FEM -1
FEM -2
`
Figure 4. Simulation result for km 17 + 320 on the Saigon East-West Highway
Approach Embankment of Can Tho Bridge
Can Tho bridge on the National Highway No. 1A crosses the Hau river which is a branch
of the Mekong river to connect Can Tho City and Vinh Long province with the total length
of 15.35 km in which the main bridge is 2.72 km long and 12.63 km
approach embankment. The Can Tho bridge project is about 160 km from the South of Ho
Chi Minh City (Figure 5).
Figure 5. Study location (km 1 + 560) on the approach embankment of the Can Tho bridge
project (Google map)
Study location km 1 + 560 m
Can Tho bridge
Starting point
End point
ASEAN Engineering Journal Part C, Vol 4 No 2 (2015), ISSN 2286-8151 p.15
The cross-section of the approach embankment of the Can Tho bridge project is about
24 m wide designed for 4 lanes. The research section locates in Vinh Long province and at
km 1 + 560. The soft ground of this section was improved using PVD. Along the 50 m
subsoil profile, there are 2 layers: 28 m very soft clay layer at the top and > 23 m fine sand
layer underneath. All soil properties of the study section are given in Table 2 and were
provided by the contractors.
FD747 PVD made in Singapore with a PVD cross-section of (96.7 x 3.5 mm) was used
for installation at 29 m length, 1.0 m spacing in triangular pattern. The initial discharge
capacity of the PVD was 4451 m3/year. Figure 6 shows arrangement of the study section,
km 1 + 560, in detail.
Preloading was constructed in three major stages up to the final height of 2.56 m. The
first fill was 0.5 m and maintained at 213 days. The second stage was 216 days long at the
total fill height of 1.0 m, and the last one was 299 days. The field monitored settlement
collected via the settlement plate S2 at the centerline of the embankment (Figure 6) is shown in Figure 7.
Similarly to the first case study, the simulated settlements were also analyzed using
Hansbo’s theory (1979) (22 TCN 244-98), FEM-1, and FEM-2. FEM meshes for the FEM-1
and FEM-2 models are showed in Figure 7a and 7b, respectively. The results are plotted in
Figure 8.
The computed settlements agree well with the monitored settlement up to a fill height
of 1.0 m in more than 400 days at an accumulated field settlement of 0.3 m. However, the
simulated settlements were markedly diverging from the monitored settlement at the fill
height of 2.56 m and the subsoil has experience a settlement of 0.5 or larger.
Figure 6. A typical cross-section of the highway embankment of the Can Tho Bridge
ASEAN Engineering Journal Part C, Vol 4 No 2 (2015), ISSN 2286-8151 p.16
Table 2. Soil Properties at the Study Location km 1 + 560 on the Approach
Embankment of the Can Tho Bridge
Soil type Thick-
ness γ E v φ c kv kh
m kN/m³ kN/m2
(o) kN/m² m/day m/day
Very soft clay 27.7 16.5 1400 0.35 13.32 13.0 2.16E-04 2.59E-04
Fine loose sand 23.3 17.5 1980 0.3 15.49 20.3 1.45E-04 1.74E-04
-700
-600
-500
-400
-300
-200
-100
0
100
200
300
0 100 200 300 400 500 600 700
Time (day)
Set
tlem
ent
(mm
) S
urc
har
ge
(0.0
1m
) .
Surcharge
Monitored settlement
`
`
Figure 7. Field monitored settlement at km 1 +560 on the approach embankment of the
Can Tho bridge project
Figure 7a. A FEM-1 mesh for the initial conditions using the PLAXIS 2D software
Clo
sed c
onso
lidat
ion
boundar
y
Closed flow boundary
ASEAN Engineering Journal Part C, Vol 4 No 2 (2015), ISSN 2286-8151 p.17
Figure 7b. A FEM-2 mesh for the initial conditions using the PLAXIS 2D software
-700
-600
-500
-400
-300
-200
-100
0
100
200
300
0 100 200 300 400 500 600 700
Time (day)
Sett
lem
en
t (m
m)
Su
rch
arg
e (
0.0
1 m
) .
Surcharge
Monitored settlement
Hansbo 1979 (22TCN 244-98)
FEM -1
FEM -2
`
Figure 8. Simulated settlements to compare with the field monitored settlement at km 1 +
560 on the approach embankment of the Can Tho bridge project
Discussion
The case studies’ results of the two projects indicate that the simulated settlements
agree well with the monitored settlements when the field consolidation settlements were
less than 0.3 m (Figure 4 & 8). Ha Hoan Hy & Tran Nguyen Hoang Hung (2011)
also reported the similar results. The consolidation settlement took place at the
first stage of a surcharge which is at low fill height (e.g., < 1.0 m), that is, small
lateral pressure (e.g., < 150 kPa) affected insignificantly to the PVD (Rixner et al.
1986). Siltation may reduce the PVD
Clo
sed c
onso
lidat
ion b
oundar
y
Closed flow boundary
ASEAN Engineering Journal Part C, Vol 4 No 2 (2015), ISSN 2286-8151 p.18
discharge capacity by a factor of 2 (Bergado et al. 1996b), but qw (e.g., ~ 300 m3/year) is
still high enough for its performance (e.g., > 150 m3/year – Holtz et al. 1991). The
consolidation settlement may be influenced by smear zone effects due to PVD installation
process (Tran-Nguyen & Edil 2011, Tran-Nguyen 2010) but these simulations ignored
these effects. It is noted that all simulations were assumed that all factors affected to PVD
were neglected. In other words, assumption of ideal PVD is relevant for this stage (22 TCN
244-98 or Hansbo 1979).
The difference between the simulated settlements and field monitored settlements
became more pronounced when the field consolidation settlement was excess 0.5 m. the
biggest settlement discrepancy occurred right after the highest surcharge (e.g., 5.03 m at
km 17 + 320 on the Saigon East-West highway, and 2.56 m at km 1 + 560 on the approach
embankment of the Can Tho bridge). The percent of settlement difference up to 20%
(FEM-1) and 57% (FEM-2). The similar behaviors were reported by Ha Hoan Hy (2011)
and Ha Hoan Hy & Tran Nguyen Hoang Hung (2011).
At higher surcharges (e.g., 3 m), many factors influence the field performance of PVD,
and settlement difference is more appreciable. Duration of surcharges was more than 600
days which is long enough for infiltration or siltation taking its effect (Chai et al. 2004). At
a large field settlements (e.g., 2.2 m at km 17 + 320 on the Saigon East-West highway),
PVD may be highly deformed due to large consolidation settlement (Tran-Nguyen et al.
2010, Tran-Nguyen 2010, Ali 1991). Friction between the filter sleeve of PVD and the soil
mass surrounding the PVD may cause synchronous deformation of both the subsoil and
PVD. Ha Hoan Hy & Tran Nguyen Hoang Hung (2011) reported that PVD discharge
capacity reduced about 50% its initial discharge capacity when the field consolidation
settlement is larger than 1.5 m for the soft ground in Ho Chi Minh City. Tran-Nguyen et al.
(2010) also showed that PVD is severely deformed at large consolidation, and PVD
discharge capacity reduces remarkably when the PVD has experienced a percent settlement
of 30% or more. Lateral pressure becoming higher under higher surcharges (e.g., 3 m)
causes reduction of PVD discharge capacity (e.g., lateral pressure >= 150 kPa – Rixner et
al. 1986). This result indicates that the assumption that PVD performed ideally at large
field consolidation settlement causes over-estimate settlement, especially wrong prediction
of consolidation time.
The results show that simulations using FEM-1 (Indraratna & Redana 2000) and
Hansbo (1979) (or 22 TCN 244-98) were close together whereas FEM-2 (Chai et al. 2001)
was appreciably diverse with the FEM-1 and Hansbo (1979) at large consolidation
settlement. The FEM-2 is only suitable to simulate for projects with small field
consolidation settlement. At large field consolidation settlement, several factors affect the
field performance of PVD, and equivalent hydraulic conductivity of PVD and subsoil
suggested by Chai et al. (2001) may not be simple as Equation (7).
ConclusionsThis study conducted back analysis utilizing the field settlement data of the two complete
projects improving soft ground using PVD: the Saigon East-West highway project at km
17 + 320 and the approach embankment of the Can Tho bridge project at km 1 + 560. The
three simulation models: Hansbo (1979) (or 22 TCN 244-98), FEM-1 (Indraratna &
Redana 2000), and FEM-2 (Chai et al. 2001) were employed for this investigation. The
Mohr-Coulomb model was employed to simulate the subsoil behavior. PVD was assumed
to be ideal performance which ignores all factors that may affect the PVD. The results
suggest the following conclusion:
Assumption of ideal PVD works well when a field settlement is less than 0.3 m for
the three simulation models for the two projects.
ASEAN Engineering Journal Part C, Vol 4 No 2 (2015), ISSN 2286-8151 p.19
When a field consolidation settlement is larger than 0.5 m, the predicted settlement
is greater than the field monitored settlement. In order words, field performance of
PVD needs to take into account for influenced factors such as PVD deformation,
lateral pressure effect, smear zone effect, siltation effect, and so on.
FEM-1 and Hansbo (1979) models are appropriate for PVD simulation.
References
[1] F.H. Ali, “The flow behavior of deformed prefabricated vertical drains,” Geotextilesand Geomembrane, Vol. 10, pp. 235 – 248, 1991.
[2] D.T Bergado, R. Manivannan, and A.S. Balasubramaniam, “Proposed criteria for
discharge capacity of prefabricated vertical drains,” Geotextile and Geomembranes,
Vol. 14, pp. 481-505, 1996a.
[3] D.T Bergado, R. Manivannan, and A.S. Balasubramaniam, “Filtration criteria for
prefabricated vertical drain geotextile filter jackets in soft Bangkok clay,”Geosynthetics International, Vol. 3, No. 1, pp. 63-83, 1996b.
[4] J.-C Chai, S.-L Shen, N. Miura, and D.T. Bergado, “Simple method of modeling
PVD-improved subsoil,” Journal of geotechnical & Geoenvironmental Engineering,Vol. 127, No. 11, pp. 965-972, 2001.
[5] J.-C Chai, N. Miura, and T. Nomura, “Effect of hydraulic radius on long-term drainagecapacity of geosynthetics drains,” Geotextiles and Geomembranes, Vol. 22, pp. 3-16,2004.
[6] S. Hansbo, “Consolidation by band-shaped prefabricated drains,” Ground Engineering,Vol. 12, No. 5, pp. 16-25, 1979.
[7] S. Hansbo, “Consolidation of fine-grained soils by prefabricated drains,” In:Proceedings of the 10th international conference on Soil Mechanics and FoundationEngineering, Stockholm, Sweden, pp. 667-682, 1981.
[8] H.H. Hy, Soft Ground Improvement Using PVD Considering Effect of PVDDeformation on Discharge Capacity - Case Studies, Thesis (Master’s), Ho Chi MinhCity University of Technology, Ho Chi Minh City, Vietnam, 2011.
[9] H.H. Hy, and T.N.H. Hung, “Effect of deformed PVD on soft ground improvementusing PVD in Vietnam,” Journal of Construction, Vol. 12/2012, pp. 66-69, 2011.
[10] R.D. Holtz, M.B Jamiolkowski, R. Lancellotta, and R. Pedroni, Prefabricated VerticalDrains: Design and Performance, CIRIA Ground Engineering report: Groundimprovement, Butterworth- Heinemann Ltd., London, United Kingdom, 1991.
[11] B. Indraratna, and I.W. Redna, “Numerical modeling of vertical drains with smear zoneand well resistance installed in soft clay,” Canadian Geotechnical Journal, Vol. 37, No.1, pp. 133 – 145, 2000.
[12] Ministry of Transport, Vietnam Code for the Use of PVD to Improve Soft Ground ,22TCN 244-98, Vietnam, 1998.
[13] J.J Rixner, S.R Kraemer, and A.D. Smith, Prefabricated Vertical Drains, Vol. IEngineering Guidelines, Technical Report, FHWA/RD-86/168, Federal HighwayAdministration Report, Virginia, United States, 1986.
[14] H.-H. Tran-Nguyen, T.B. Edil, and J.A. Schneider, “Effect of deformation ofprefabricated vertical drain (PVD) on discharge capacity and the characteristics of PVDsmear zone,” Geosynthetics International, Vol. 17, pp. 431-442, 2010.
[15] H.-H. Tran-Nguyen, and T.B. Edil, “The characteristics of PVD smear zone,” Advancedin Geotechnical Engineering International Conference, Geotechnical SpecialPublication, Dallas, Texas, United States, pp. 748-757, 2011.
ASEAN Engineering Journal Part C, Vol 4 No 2 (2015), ISSN 2286-8151 p.20